Download UPGRADING THE CAMPUS NETWORK TO LAYER-3 Rizwan Ahmad

Rizwan Ahmad
UPGRADING THE CAMPUS
NETWORK TO LAYER-3
Faculty of Information Technology
2011
VAASAN AMMATTIKORKEAKOULU
UNIVERSITY OF APPLIED SCIENCES
Degree Programme in Information Technology
ABSTRACT
Author
Title
Year
Language
Pages
Name of Supervisor
Rizwan Ahmad
Upgrading the Campus Network to Layer-3
2011
English
46
Antti Virtanen
The aim of this project is to upgrade the campus network of Vaasan ammattikorkeakoulu (VAMK), University of Applied Sciences. This project is divided
into two parts, configuration of the VLAN subnets and configuration of the network routing.
This thesis focuses on configuration of VLANs, configuration of DHCP scopes
for subnets, configuration and implementation of network routing such as static
and dynamic routing (RIP). The theoretical part explains the basic network fundamentals and provides enough comprehensive background about the network
routing. A major part of theory is devoted to the explanation of VLAN concept
and routing protocols.
During the implementation, VLAN subnets are created. DHCP scopes are configured for each subnet to get IP addresses and finally static and dynamic routing is
configured to fulfil the network demands. After the completion of implementation, tests have been made to verify the practical work. All the tests were successfully completed and a detailed report is in the results & analysis.
After the implementation, VAMK campus network is changed to Layer 3 network. Even access layer switches are configured to do the routing. Implementation of the routing at the edge of the network provides greater security, scalability,
control and bandwidth conservation.
Keywords
IP Routing, Static Routing, Dynamic Routing, RIP Routing,
VLAN, IEEE802.1Q
1
CONTENTS
ABSTRACT
1
INTRODUCTION ............................................................................................ 8
2
BACKGROUND & GOALS OF THE PROJECT ........................................... 9
2.1 Background Information ........................................................................... 9
2.2 Working environment of the project ......................................................... 9
2.3 Purpose of the project ............................................................................. 10
2.4 Objectives of the project ......................................................................... 10
3
PROJECT DESCRIPTION ............................................................................ 11
3.1 IP Routing ............................................................................................... 11
3.1.1 Static Routing .............................................................................. 11
3.1.2 Dynamic Routing ........................................................................ 11
3.1.3 Routing Information Protocol (RIP) Ver.1 ................................. 12
3.1.4 Routing Information Protocol (RIP) Ver.2 ................................. 13
3.2 Virtual LANs .......................................................................................... 14
3.2.1 Static VLANs .............................................................................. 16
3.2.2 Dynamic VLANs ........................................................................ 18
3.2.3 Tagged VLANs ........................................................................... 18
3.3 Dynamic Host Configuration Protocol (DHCP) ..................................... 18
4
IMPLEMENTATION OF THE PROJECT.................................................... 20
4.1 Overview of the Network........................................................................ 20
4.2 Overview of the Design .......................................................................... 21
4.3 Implementation of the Project ................................................................. 22
4.4 Cloning switch configuration.................................................................. 22
4.5 Replacing Access Layer Switches .......................................................... 23
4.6 Configuring DHCP ................................................................................. 24
4.7 Configuring VLANs ............................................................................... 25
4.7.1 Menu Interface ............................................................................ 25
4.7.2 CLI Interface ............................................................................... 28
4.8 Configuring IP routing ............................................................................ 30
4.8.1 Configuring IP Static Routes ...................................................... 31
2
4.8.2 Configuring Routing Information Protocol (RIP) ....................... 32
5
TEST, RESULTS AND ANALYSIS ............................................................. 38
5.1 Test of the VLAN ................................................................................... 39
5.2 Test of the Static Routing ....................................................................... 42
5.3 Test of dynamic routing .......................................................................... 43
6
CONCLUSION .............................................................................................. 45
REFERENCES...................................................................................................... 46
3
ACKNOWLEDGEMENT
I would like to dedicate this thesis to my dearest parents Mansoor Ahmad Dogar
(father) and Mubashra Mansoor (mother).
First of all, I would like to thank Almighty God, the most merciful, the most beneficent for his guidance and blessings in making this thesis successful.
I would like to express my deep gratitude to my supervisor Antti Virtanen for his
detailed and constructive comments, feedback and for his important support
throughout this work.
I am heartily grateful to Hannu Teulahti, VAMK Linux Administrator, who was
always helpful by all the means to complete this project.
I would like to express my special thanks to my parents for their help, support,
love and prayers to complete my studies at this institution.
Finally, I am thankful to my friends for their help and everlasting support during
my studies in Vaasa.
Vaasa, 29th April 2011.
Rizwan Ahmad
4
ABBREVIATIONS
ARP
Address Resolution Protocol
ARPANET
Advanced Research Projects Agency Network
BOOTP
Bootstrap Protocol
CIDR
Classless Inter-Domain Routing
CLI
Command Line Interface
DHCP
Dynamic Host Configuration Protocol
EIGRP
Enhanced Interior Routing Protocol
ICMP
Internet Control Message Protocol
IEEE
Institute of Electrical and Electronics Engineers
IP
Internet Protocol
IPTV
Internet Protocol Television
IPv4
Internet Protocol Version 4
IPv6
Internet Protocol Version 6
IRDP
ICMP Router Discovery Protocol
IS-IS
Integrated Intermediate System To Intermediate System
LAN
Local Area Network
MAC
Media Access Control
OSPF
Open Shortest Path First
QoS
Quality of Service
5
RFC
Request for Comments
RIP
Routing Information Protocol
SNTP
Simple Network Time Protocol
SNMP
Simple Network Management Protocol
TCP
Transmission Control Protocol
VAMK
Vaasan ammattikorkeakoulu
VLAN
Virtual Local Area Network
VLSM
Variable Length Subnet Mask
VoIP
Voice over Internet Protocol
6
LISTS OF PICTURES, GRAPHS AND TABLES
Figure 1.
RIP V1 message format
p.13
Figure 2.
RIP V2. message format
p.14
Figure 3.
VLAN segmentation
p.15
Figure 4.
Protocol base VLAN support
p.17
Figure 5.
DHCP client server procedure
p.19
Figure 6.
VAMK network design before implementation
p.21
Figure 7.
HP Procurve Switch Configuration
p.23
Figure 8.
HP Procurve & Cisco Switch
p.24
Figure 9.
DHCP subnet scopes
p.25
Figure 10.
HP Procurve VLAN Menu
p.26
Figure 11.
HP Procurve VLAN support menu
p.26
Figure 12.
VLAN creation menu
p.27
Figure 13.
VLAN name menu
p.27
Figure 14.
VLAN port assignment
p.28
Figure 15.
HP Procurve CLI interface
p.29
Figure 16.
Management & data VLAN
p.30
Figure 17.
Static route configuration
p.31
Figure 18.
Default route configuration
p.32
Figure 19.
RIP Parameters
p.34
7
Figure 20.
Enabling RIP Routing
p.34
Figure 21.
Disabling RIP Routing
p.35
Figure 22.
RIP configuration on VLAN interface
p.35
Figure 23.
RIP redistribution
p.36
Figure 24.
Redistribute static and connected routes
p.36
Figure 25.
Changing the cost of a route
p.36
Figure 26.
Loop Prevention Method
p.37
Figure 27.
Network structure after the implementation
p.38
Figure 28.
Testing VLAN
p.40
Figure 29.
IP configuration of workstation
p.41
Figure 30.
Testing Static routing
p.42
Figure 31.
Testing RIP routing
p.43
8
1
INTRODUCTION
The aim of this project is to upgrade campus network for supporting new technologies, provide greater security, high data connectivity. The rapidly growing
demand for high data rate communication system requires network upgrading
with time. The goal is to upgrade campus Layer 2 network into Layer 3 network
architecture, to provide high speed data connectivity, reliable and efficient network performance. Layer 3 architecture limits the broadcast domains, is easier to
manage, allows greater scalability, supports routing protocols, IP services, bandwidth usage, software tools to check network performance troubleshooting and
configuration. /10/
Nowadays, everyone is familiar with computer and internet. Communication is a
major activity around everyone’s daily life. Besides, the new and innovative technologies are arising to fulfil the daily varying demands. Each network requires
high bandwidth, secure and fast communication for network to provide greater
Quality of Services to delay-sensitive devices or applications such as VoIP, IPTV
and video conferencing. Since the demands for real time applications (VoIP,
IPTV) are increasing, it is significant to make network not only respond quickly
but needs to provide Quality of Services (QoS) as well.
This project will describe the configuration of the Layer 3 network framework
implementation, basic configuration of routing switch HP Procurve 2910-al, creating, assigning and configuring VLANs for each subnet, configuring IP routing
and DHCP server to allocate IP addresses for new subnets.
9
2
2.1
BACKGROUND & GOALS OF THE PROJECT
Background Information
Now, the world is in a new era, rapidly growing and high demands with many different types of networks have directed the implementation of the campus network
into a new design. The new design is scalable and meets all the current demands
of the network requirements.
Before this project, Vaasa University of Applied Sciences had Layer 2 (Data-Link
Layer) network, all the routing has been done on the main routing switch. Each
VLAN traffic forwards toward the routing switch for routing and subnet assignments that makes router performance lower and increases the bandwidth usage.
The Layer 2 switches in the classrooms did not support routing and configured to
forward all the traffic that needs to be routed towards the routing switch.
2.2
Working environment of the project
This project was implemented in Vaasa University of Applied Sciences. Vaasa
University of Applied Sciences located at Vaasa, Finland (Western Finland).
Vaasa is one of the largest student cities with more than 12000 students. Vaasa
provides work for a diversified range of experts as well. There are many employers in Vaasa like Wärtsilä, ABB and Vacon.
University of Applied Sciences, Vaasan ammattikorkeakoulu (VAMK), started its
operation on August 1, 1996 with the degree programs in Technology and Communication, Business Economics and Tourism and Health Care and Social Services. VAMK has collaborative universities all over the world. In addition to the
international co-operation mainly in Europe, the student exchange has also been
active. VAMK has 350 foreign degree and exchange students from about 40 countries. /2/
VAMK has two campuses, Wolffintie and Raastuvankatu. The Wolffintie campus
provides the Information Technology studies and Raastuvankatu campus provides
10
International business, Catering and nursing studies. The project work has done in
Wolffintie campus.
2.3
Purpose of the project
Today’s technologies are developing day by day. Organizations are upgrading
their networks and services to adopt new technologies. Organizations need a single network infrastructure that can provide all the demanding services like high
data transmission, greater security, scalability, supporting for new technologies,
routing protocols and other IP services instead of many different networks. All the
companies want to reduce their cost and efforts. Since the demand for these kind
of applications are increasing, it is vital that a better communication network is
implemented, which provides the solution for real time applications and take care
of data flow as well. This project will produce a network with faster data transfer
speeds, reliability, increased security and the ability to meet the future demands of
the campus community.
2.4
Objectives of the project
This project has the following objectives:
 Higher data transfer speeds (up to 1GigaBit to each workstation)
 Minimize broadcast domains
 IP Routing
 Static Routing
 RIP Routing
 Configuration of VLANs
 Configuration of DHCP
11
3
PROJECT DESCRIPTION
3.1
IP Routing
The IP routing in the IP network is the set of tasks required to move an IP packet
from router to router to its destination, as specified in the IP header field /4, 1-2/.
IP routing is the most important task in an IP network. Two types of IP routing
exist in an IP network:
 Static Routing
 Dynamic Routing
3.1.1
Static Routing
The most basic and simplest routing method is called static routing. In static routing, the routing table is constructed manually. All the routes are fixed and network
administrator has to manually create, modify or delete route from the routing table
whenever network changes occurred. Well, static routing is easy to configure,
gives more control for network administrator to control the network and the network path between two destinations always known precisely. But static routing is
used only in small network to get better performance. /5, 317-319/
3.1.2
Dynamic Routing
Dynamic routing performs the same function as static routing except it is more
robust. In dynamic routing, routing tables are created automatically in such a way
that adjacent routers exchange routes with each other. Each router calculates its
own best route using metrics and rules. The selection of best route is based on the
metrics such as number of hops, delay, bandwidth, link cost and these values are
updated by using protocols. Dynamic routing protocols can change the routing
table in the router if possible routes are changed. Dynamic routing has capability
to update the routing table automatically, so it does not consume the network administrator time but if it’s not configured properly, it can create diverse problems
such as routing loops. Dynamic routing is supported by several protocols: /5, 320/
12
 Routing Information Protocol (RIP)Ver.1
 Routing Information Protocol (RIP)Ver.2
 Open Shortest Path First (OSPF)
 Enhanced Interior Gateway Routing Protocol (EIGP)
 Integrated Intermediate System To Intermediate System(IS-IS)
Since this thesis is only focusing in Routing Information Protocol (RIP), we will
not focus on other protocols.
3.1.3
Routing Information Protocol (RIP) Ver.1
Routing Information Protocol (RIP) was originally used by ARPANET and was
designed to work with small sized networks. RIP ver.1 is a classful, distance vector routing protocol that does not support CIDR (Classless Inter-Domain Routing)
and VLSM (Variable Length Subnet Mask). RIP ver.1 is defined in RFC 1058. It
is using hop count as a metric. Routes with a hop count more than 15 are unreachable. RIP ver.1 updates are broadcasted every 30 seconds. RIP ver.1 uses split horizon method with poison reverse. RIP ver.1 message format is shown below in
the figure. In the figure, three fields are specified in the four byte header shown in
grey. The command field specifies the message type, the value of command field
1 is used for a request and 2 is for a reply. Version field shows the protocol version, 1 for RIP ver.1 and 2 for RIP ver.2. The third field is labelled “Must be
Zero” This field provide space for future expansion of the protocol. Next field is
“Family Address Identifier”, the value of this field set to 2 for IP unless a router is
requesting a full routing table, in which case the field is set to zero. IP Address
field carry the address of the destination route, which maybe a network, subnet, or
host address. Metric field is used to hop count between 1 and 16. Sending router
increases the metric before sending out message. /1/
13
Figure 1. RIP V1 message format /1/
3.1.4
Routing Information Protocol (RIP) Ver.2
Routing information protocol (RIP) was originally used by ARPANET and was
designed to work with small-sized network /1/. RIP ver.2 is defined in RFC 1723.
RIP ver.2 has the same message format as RIP ver.1 but two significant extensions are added:
The first extension in RIP ver.2 message format is the support of VLSM that allows 32 bit mask to be included in the RIP route entry. The receiving router no
longer depends upon the subnet mask of the inbound interface or the classful mask
when determine the subnet mask of the route. /3/
The second extension in RIP ver.2 message format is the addition of the next hop
address. The purpose of this extension is to find the better next hop address if exists /3/. RIP Ver.2 message format is below in the figure:
14
Figure 2. RIP V2. Message format /3/
In RIP V2, three fields (Route Tag, Subnet Mask, Next Hop) are added into RIP
message. Route tag field provides a tag for distinguishing between internal routes
and external routes, Subnet field contains the subnet information of the destination
because of adding this field RIP V2 supports VLSM and Next Hop field indicates
the IP address of the next hop where packets are sending to reach the destination.
3.2
Virtual LANs
Logically divided physical network into different IP sub network called Virtual
LAN. In Virtual LAN, we can group user logically instead of physical locations.
Using VLANs, it can control high bandwidth usage in our network to group high
bandwidth users to switch on low-traffic segments. It can also organize users from
different Virtual LAN segments according to their need for common resources.
Virtual LANS provides many benefits as follows: /6/
 Security ( groups that have sensitive data can be separated from the network)
 Increased Performance (Creates layer 2 logical networks reduced broadcast traffic and boost the performance.)
 Cost Reduction (More efficient use of network bandwidth, uplinks and
less need for expensive hardware upgrades)
15
 Broadcast Storm Mitigation (Virtual LANS reduces the number of devices
that can participate in the broadcast storm)
 Improve Traffic Control (We can improve traffic control at the edge of our
network by separating different protocol traffic)
 Improved IT Staff Efficiency(All the policies and permissions are configured for a particular VLAN, All the policies are applied when provision a
new switch and assign the VLAN)
Figure 3. VLAN segmentation
In figure 3, LAN and VLAN segmentation is shown. In traditional LAN segmentation all the segments are in the same broadcast domain and it reduces the efficiency of the network. In VLAN segmentation, broadcast domains are divided
into the number of VLANs, as shown in the figure, three broadcast domains are
exists, physically the members of VLAN domains are in the different locations
but logically they are in the same broadcast domain.
In this HP Procurve 2910al switch we have three types of VLANs,
 Static VLANs
 Dynamic VLANs
 Tagged VLANs (Trunk VLAN)
16
3.2.1
Static VLANs
In static VLAN, ports are manually assigned to a VLAN. Data packets are bridged
from source to destination port in same VLAN and cross domain broadcast packets are eliminated to save bandwidth in the switch. Static VLANs can be configured with CLI interface or Procurve menu interface. Static VLANs are configured
with a VLAN ID, VLAN name and port members. Further static VLANs divided
into many types as follows: /7, 2-5/
 Port-Based VLANs
 Protocol-Based VLANs
 Designated VLANs
Tagged: A port that complies with the 802.1Q standard, including priority settings
and allows a port to join multiple VLANs.
Untagged: The switch allows no more than one untagged VLAN assignment per
port. If a port is configured only untagged, it does not use or forward 802.1Q
VLAN tagging.
Port-Based VLANs
Port-Based VLAN creates a layer-2 broadcast domain. This layer-2 broadcast domain comprised of member ports that bridge IPv4 traffic among themselves. In
Procurve 2910al, a port can be a member of one untagged Port-based VLAN and
all other Port-Based VLAN assignments for that port must be tagged. Port-based
VLANs can be configured without IP address or with unique IP address. We can
use multiple IP address to create multiple subnets in the same port-based VLAN
as well. Procurve 2910al switch can route traffic between port-based VLANs,
Port-based and Protocol-based (IPV4) if the IP routing is enabled on the switched.
/7, 2-7/
17
Protocol-Based VLANs
Protocol-based VLANs are creating a layer-3 broadcast domain for a specific protocol. In Procurve 2910al, a port’s untagged protocol-based VLAN membership
can include up to four different protocol types. It means that a port can be an
untagged member of one of the following:
-Four single-protocol VLANs
- Two protocol-based VLANs where one includes a single protocol and other includes up to three protocols
- One protocol-based VLAN where VLAN includes four protocols
A protocol-based VLAN port can be tagged member of any protocol-based
VLAN. If IP routing is enabled on switch, switch can internally route the traffic
between multiple protocol-based (IPV4) VLANs and port-based VLANs. We
need to have an external router to route traffic between other protocol-based
VLANs. Please see the table below for more routing information on Procurve
2910al switch. /7, 2-8/
Figure 4. Protocol based VLAN support. /7, 2-10/
Designated VLANs
These designated VLANs are also Port-based VLANs but these VLANs used for
special management purposes. These VLANs separate management traffic from
18
network traffic and provide improved security and availability for management
traffic. In Procurve 2910al, these are designated VLANs: /7, 2-5/
 The Default VLAN
 The Primary VLAN
 The Secure Management VLAN
 Voice VLANs
3.2.2
Dynamic VLANs
Dynamic VLAN is not used in this thesis work so it will be explained briefly.
However, it is useful to know what a dynamic VLAN is. In dynamic VLAN, we
can assign switch ports to VLAN dynamically. Dynamic VLAN assignment is
based on the source MAC address of the device connected to the port. Dynamic
VLAN is configured with a special server that has a database of all devices MAC
addresses to assign VLANs dynamically. If a host moved from one switch port to
another switch port in the network, the switch dynamically assign the proper
VLAN to the host. /7, 3-3/
3.2.3
Tagged VLANs
Tagged VLAN is point-to-point link between two network devices that carries
more than one VLAN in the network. We can extend the VLANs in the entire
network with tagged. HP Procurve 2910al supports IEEE 802.1Q for coordinating
tagged ports on gigabit Ethernet interfaces. Tagged ports are useful in network
backbone, where many VLANs traffic is handled. In Cisco terms we called tagged
VLAN as trunk VLAN. /7, 2-41/
3.3
Dynamic Host Configuration Protocol (DHCP)
Dynamic Host Configuration Protocol (DHCP) is an Internet protocol that enables
a server to dynamically allocate and distribute IP addresses and additional TCP/IP
configuration. RFCs 2131 and 2132 define DHCP as an Internet Engineering Task
19
Force (IETF) standard based on Bootstrap Protocol (BOOTP). DHCP uses a client/server procedure (see Figure below), where DHCP server receives a
DHCPDiscover message for an IP address from a client when it first attempts to
the network. After receiving the DHCPDiscover message, DHCP server broadcast
DHCPOffer message if server has an IP address configuration to offer to the
client. DHCPOffer message contains an unleased IP address, subnet mask, default
gateway and other TCP/IP configuration. After DHCP client selects the DHCPOffer, it broadcast the DHCPRequest message. The message contains the IP address
from the DHCPOffer that it selected. If the client is renewing to a previous lease,
this packet might be unicast to the server. Finally DHCP server acknowledged to
DHCPRequest message by broadcast DHCPACK message. After receiving an IP
request, the DHCP server assigns it an IP address and updates its database table,
noting that which client has the address and expiration time of the address. When
the time expires, the DHCP client needs to renew the amount of time or negotiate
a new amount of time for a new IP address. /8/
Before the advent of DHCP, most of TCP/IP configurations were maintained
statically. Each individual host or workstation needs to be configured manually IP
address, subnet mask, default gateway as well as other TCP/IP parameters. Configuring and maintaining static TCP/IP parameters for multiple work stations and
network devices might be a difficult task, especially in a large network. DHCP has
solved this problem; Using DHCP an administrator can make changes in the Client’s IP configuration without visiting each and every client. The user at the
workstation only needs to release and renew their DHCP lease.
Figure 5. DHCP client server procedure/9/
20
4
4.1
IMPLEMENTATION OF THE PROJECT
Overview of the Network
VAMK network has basically been divided into two parts, VAMK Staff network
and VAMK classroom network. VAMK staff is entirely separate network for
staff, office and other management tasks. VAMK classroom network is for students. All the security policies, restrictions, access rights and personal storage
quota are applied on this network. The users of this network have limited access to
the network resources as configured by the network administrator.
This thesis focuses on the classroom network. Before the implementation of this
thesis, VAMK had Layer-2 switching network. All the routing had to been done
on the central routing switch. Classrooms were not configured as own
VLANs/subnets, but at least two or three classrooms share the same subnet. All
the trunk/tagged traffic was sent to the trunk port to reach its destination that increased the network traffic and more bandwidth usage, because the access layer
switch could not able to route the traffic to its destination. See the figure below for
more information. In the figure, a Layer-2 network is shown. In this network,
802.1Q VLANs are used to segment network users and broadcast domains. Because of Layer-2 switch limitation no communication is allowed directly between
VLANs. All the Inter VLAN communication is done by the routing switch. Routing switch is handling all the Layer-3 traffic for the whole network such as forward DHCP queries to DHCP server.
21
Figure 6. VAMK network design before implementation
4.2
Overview of the Design
The design was started by getting the current running configuration of the Cisco
access layers, distribution layer and core layer switches/routers, because it helps to
understand, how the network is configured and working.
The whole network was working and students were using it. It was not allowed to
interrupt any classroom network because of the configuration. So discussion with
network administrator was arranged. I was suggested that firstly, same configure
on new HP switches should be made, because it keeps the network running. Later,
the desired configuration changing according to the demands can be made.
VAMK has already Dynamic Host Configuration Protocol (DHCP) Server to allocate IP address to the clients/subnet. There was no need to configure new DHCP
server, same DHCP server was used to allocate IP addresses to new VLAN subnets. Currently installed Cisco access layer switches do not support routing. The
22
network routing was not configured on the new HP switches yet, because it does
not affect the network now.
4.3
Implementation of the Project
The project implementation is subdivided into five parts:
 Cloning switch configuration
 Replacing the access layer switch
 Configuring DHCP
 Configuring VLANs
 Configuring routing
4.4
Cloning switch configuration
This thesis consists of a real time working network. It was not affordable to lose
the network connectivity. After discussion with network administrator, a decision
has been made that first HP Procurve switches should configure with the same
configuration as Cisco Layer-2 switches, which will keep network running without interruption. Furthermore, configuration could be changed later according to
the demands.
For the cloning of the switches configuration, we have started to read the configuration from old Cisco Catalyst switches (Layer 2) and configured new HP Procurve switches with the same configuration (Layer 2). After implementing the
(Layer 2) configuration, still all the routing has been done on main routing switch,
the routing feature of HP Procurve switch was not used yet. Because of two different hardware manufacturers, the command syntax and configuration interface
was different. For more understanding, please see the configuration syntax in the
figure below:
23
Figure 7. HP Procurve Switch Configuration
4.5
Replacing Access Layer Switches
Once the cloning of the switches configuration has been done, It was time to replace the switches but to make sure that the correct configuration has been made
on the switches; a test has been done before replaced the switches in the network.
In some places, I could not fix the new HP Procurve switches but this problem
was solved with the help of our professional technician. Please see the figure below to have a look of switches:
24
Figure 8. HP Procurve & Cisco Switch
4.6
Configuring DHCP
In the previous chapter, Dynamic Host Configuration Protocol (DHCP) is explained in detail. In this topic, the configuration of DHCP is described. Well,
VAMK have had running DHCP server in the network; there was no need to build
a new DHCP server. After replacing the switches, we had not used any feature of
new hardware yet; network was running with the same network configuration like
earlier. New DHCP subnet scopes have been created for each class room. It was
required to separate each class room network. This DHCP server was already configured and running in the VAMK network, new IP scopes were created to fulfil
our requirements. In the figure below, DHCP scope configuration is shown:
25
Figure 9. DHCP subnet scopes
4.7
Configuring VLANs
The definition, types and purposes of VLANs are explained in previous chapter.
In this topic, implementation and configuration of port-based VLANs and tagged
VLANs are explained because this thesis focuses only on these VLANs. In this
HP Procurve 2910al switch, we have two ways to configure VLANs:
 Menu Interface
 CLI Interface
4.7.1
Menu Interface
Port-based VLANs can configure with menu interface. Menu interface is simple
and easy to use for beginners. In the factory default state, all ports are assigned to
default VLAN. Also in the factory default, switch is enabled to support 256
26
VLANs but it is possible to reconfigure the switch to support up to 2048 VLANs.
The figures below shows, how to configure the switch for supporting more
VLANs, Adding/Editing VLAN names and creating new VLANs.
Figure 10. HP Procurve VLAN Menu
After get into the VLAN configuration menu, there are three more submenu options for different VLAN configurations.
 VLAN Support menu is for activate and configure or deactivate VLAN
support. Screen shot is below for more details.
Figure 11. HP Procurve VLAN support menu
27
 VLAN Name menu is for configures names for VLANs for grouping
switch ports. VLAN menu can add, edit and delete VLANs as well in this
menu. Screen shots are below for better understanding.
Figure 12. VLAN creation menu
Remember to save the configuration before exiting the submenu. After having
saved the configuration, we will see the new VLAN name and ID in the VLAN
names menu. See the screen shot below:
Figure 13. VLAN name menu
 VLAN port assignment menu is used to assigned switch ports to VLANs.
Switch allows no more than one untagged VLAN assignment per port. By
default, all ports are untagged members of default VLAN. A switch port
can be a tagged member of multiple VLANs.
28
Figure 14. VLAN port assignment
4.7.2
CLI Interface
CLI stands for Command Line Interface. CLI interface is powerful tool and gives
more control to manage the switch. Actually, it is the real interface to configure
the switch. The configuration of both port-based and protocol-based VLANs can
be done with CLI. Each switch port should be at least an untagged member of a
VLAN. By default, all ports are untagged members of default VLAN. There are
many features of this switch that can only be configured with CLI. The figure below shows the configuration with CLI commands:
29
Figure 15. HP Procurve CLI interface
In the following figure, data VLAN and management VLAN are shown, data
VLAN is configured to carry only user-generated traffic. Management VLAN is
configured to access the management capabilities of the switch. A switch can be
managed via Telnet, SSH or SNMP.
30
Figure 16. Management & data VLAN
4.8
Configuring IP routing
IP routing is the process of moving IP packets from source to destination. The
routing decision is made by a router which provides the physical connection between the networks. Such router must be configured with routing protocols to established communication beyond the local network.
Basic IP routing configuration consists of enabling IP routing, configure IP addresses, static routes and dynamic routing protocols. Such as Routing Information
Protocol (RIP). HP Procurve 2910al offers the following IP routing features:
 IP Static Route
 Routing Information Protocol (RIP)
31
 ICMP Router Discovery Protocol (IRDP)
 DHCP Relay
This document will describe the configuration of first two IP routing features because only these features are used in this project:
4.8.1
Configuring IP Static Routes
HP Procurve 2910al supports up to 256 static routes. Static routes can be added
directly to the routing table. Static IP routes remain in the routing table as long as
the IP interface of next hop router is up. These specific parameter needs to configure a static route:
 IP address and destination mask for route’s destination network/host.
 The route path which can be the following: IP address of a next hop router
or a null interface to drop the traffic. If the route path is a null interface, all
the traffic forward to this interface will be dropped by the router.
The figure below shows the command to create a static route entry in the routing
table.
Figure 17. Static route configuration
32
Default route is another static useful route. All the traffic that has destination unreachable through any other IP routing table, sends to this default route interface.
Please see the figure below for more information:
Figure 18. Default route configuration
4.8.2
Configuring Routing Information Protocol (RIP)
Routing Information Protocol (RIP) is the oldest of the distance vector routing
protocols. Detailed explanation is in the last chapter, now the configuration of RIP
will be described in this topic. HP Procurve 2910al supports the following RIP
types:
 Version 1
 Version 2 (Default)
 V1 compatible with V2
Before the configuration of RIP, I like to explain the RIP global parameters and
RIP interface parameters for better understanding.
In HP Procurve 2910al switch, there are four RIP global parameters: RIP State,
auto summary, metric and redistribution. RIP state parameter is used to configure
the RIP version. By default, “RIP V2 only” is disabled and switch is supporting
RIP V1 & V2 in global configuration, Advertisement of auto summarized routes
33
is enabled, default metric for imported routes is 1 and redistribute of static & connected routes is disabled but RIP redistributes connected routes by default, when
RIP is enabled.
RIP interface parameters are more useful. It can be configured differently according to network demands. There are six RIP interface parameters: RIP version,
metric, IP address, loop prevention, receive and send. RIP version parameter is
used to configure the RIP version on the interface level. By default, this parameter
value is V2 only. Metric parameter is used to add the cost of an imported route
learned on the interface. IP address parameter is used to control the routes which
are learned or advertised. Loop prevention parameter is an important parameter
because it protects the network from routing loops. Receive parameter is used to
define the RIP version for incoming packets and send parameter is used to define
the RIP version for outgoing packets. The following figure shows the default values of RIP global and RIP interface parameters:
34
Figure 19. RIP Parameters
The following configuration shows how to enable RIP routing on a switch. IP
routing should be enabled prior to enabling RIP.
Figure 20. Enabling RIP Routing
To disable RIP, the following command is used:
35
Figure 21. Disabling RIP Routing
The most useful RIP parameters are explained below with command syntax and
figures:
 RIP on VLAN interface
Configuration of RIP protocol on a VLAN interface is an important and useful
task. RIP can be configured on a specific or all VLAN IP addresses. By default,
RIP V2 is used but we can always change it to individual VLAN interface basis.
To enable RIP on VLAN context, see the figure below for configuration commands:
Figure 22. RIP configuration on VLAN interface
 RIP Redistribution
HP Procurve 2910al only supports static and connected route redistribution. Redistribution of any other routing protocol is not supported. Switch is using RIP to
advertise redistributed routes to its RIP neighbours. Redistribution filters can be
used to get more control on redistribution of routes. These filters permit or deny
redistribution for a route based on interface or destination network address. Figure
below shows the configuration that how to filter out the static/connected routes on
network 192.168.2.0:
36
Figure 23. RIP redistribution
For enabling the redistribution of static/connected routes into RIP, the command
is shown in figure below:
Figure 24. Redistribute static and connected routes
 Cost of routes on VLAN
By default, the switch increases the cost by adding one to the route metric which
is learned by RIP interfaces. In RIP, a route that has metric 16 considered unreachable. We can configure a specific interface to increase RIP routes metric.
Figure below shows the configuration command to increase the cost of the route
by increasing metric:
Figure 25. Changing the cost of a route
37
 Root Loop Prevention Method
HP Procurve 2910al switch, RIP uses two methods to prevent routing loops. By
default, poison reverse is the default loop prevent method. It basically assigns
route cost of 16 before to advertise it to the same interface from the routing switch
has been learned the route. Second method is called split horizon. The basic rule
of this method is that the routing switch does not advertise a route to the interface
from the routing switch learned the route. Routing loop prevention methods can
be configured on individual VLAN interfaces. To disable a loop prevention
method automatically enables the other one. Figure below shows the configuration
command:
Figure 26. Loop Prevention Method
38
5
TEST, RESULTS AND ANALYSIS
After the implementation, I could say that a Layer 3 network is build and working
properly. Static and dynamic (RIP) routing is enabled on each HP 2910 switch
that makes the network to a Layer 3 network and many benefits are realized.
These benefits include IP subnetting, greater network security, availability
through high speed data connectivity, Layer-3 protocol filtering and intercommunication between Layer-2 VLANs. Now each edge switch is capable of doing IP
routing and by limiting to send all the data packets towards the main router to
route them, it increased bandwidth conservation. Dynamic routing is configured
and working properly that makes the IP routing more efficient. The following figure shows the network structure after the implementation of the project.
Figure 27. Network structure after the implementation
39
After the completion of implementation tasks, now a test should be done to make
sure that network is working properly. A methodically test has been done to easily
find any problem if it exists. The test is divided into three parts:
 Test of the VLAN
 Test of static routing
 Test of dynamic routing
5.1
Test of the VLAN
First VLAN test has been done to make sure that the basic configuration is working properly. In VLAN testing we will make sure that each subnet is getting
proper IP addresses and other configuration parameters.
This test has been done in a classroom network. I mentioned earlier that each
classroom has its own sub network. All the workstation in the classroom are connected with HP Procurve access switch and configured to be a member of same
VLAN. For VLAN test, classroom workstation is used that is connected to the
switch with network cable. After successfully login to the workstation, IP configuration of the workstation has been checked to verify that it is getting the
proper IP configuration from DHCP server. The figure below shows the test made
on the switch:
40
Figure 28. Testing VLAN
In figure 28, the switch displays the ports status and MAC addresses of the workstations that are connected on the VLAN. The workstation is connected to the
switch port 10 which we are using to test the VLAN. As you can see in the figure,
the port status is up and the MAC address is registered in the VLAN with port
number. Now IP configuration of the workstation has been checked to verify if the
VLAN is working properly. The figure below shows the workstation IP configuration:
41
Figure 29. IP configuration of workstation
Figure 29, shows the IP assigned to the workstation. This IP address is from the
subnet 192.168.86.0/25 because the specific switch port is belongs to VLAN 405
and VLAN 405 is configured to get the IP addresses from the subnet
192.168.86.0/25. This is the proper IP address as we configured to assign for this
VLAN. Physical address of this network adapter is the same which we have seen
in the switch VLAN MAC-addresses. I tried to access the internet and other network resources such as intranet, local student shares, staff shares and personal
disk drive. All the network resources are accessible which means that the VLAN
is working.
42
5.2
Test of the Static Routing
Static routing is the simplest form of network routing. Static routing is usually
used in small networks to get better performance and static routing could be used
in larger network for some specific routes. Because the simplicity of the static
routing, it was easy to configure and test it. For the testing of static routing, a routing enable switch has selected which has static routes as well and tried to trace the
route to the destination network. For this purpose, traceroute command is used of
the routing switch. The figure below shows the test results:
Figure 30. Testing Static routing
Figure 30, shows the static routes in the routing switch and when routing switch
tried to trace the routes, it shows the next hop address which is configured in the
static route. It means the static routing is configured and working properly.
43
5.3
Test of dynamic routing
Dynamic routing is using many routing protocols but only RIP V2 is used in this
project work. For testing the RIP routing, a routing switch is selected, which is
configured for RIP routing. Figure below shows the results:
Figure 31. Testing RIP routing
44
Figure 31, shows the RIP routing table of the routing switch. There are many RIP
routes and two routes are chosen from the routing table for testing. RIP routing
test has made with traceroute command; routing switch has done the routing as it
is configured in the routing table. The figure 31 shows the RIP routing test result
and the result clearly shows that the RIP routing is working properly.
45
6
CONCLUSION
In this project, a Layer 2 (Data Link Layer) switching network has been changed
to Layer 3 (Network Layer) routing network. The required objective of this project has been achieved within the time frame. In this project work, static routing,
dynamic routing, IP subnetting, DHCP configuration, IP configuration, VLAN
configuration and intercommunication between Layer 2 VLANs have been done.
This project provides an opportunity to implement the theoretical knowledge of
data transmission and packet switched network. Before this project, I was familiar
with Cisco switch configuration but we have used HP Procurve switches in this
network project. It was a completely new task for me to configure HP switches
because of its different command syntax and features. During this project, I have
created 20 new IP subnets to DHCP, Approximately 14 new VLANs are created
and configured, some VLANs are reconfigured as well, around 25 static/default
routes were configured and approximately 30 RIP interfaces were configured to
advertises and learn new RIP routes. I have learned many things which could
never be learnt by theory without practice.
After the implementation of this whole project, I conclude that:
Layer 3 network provides scalability via routing support, WAN interfaces, and
support of network layer protocol, greater security via IP subnetting and IP filtering, availability through high speed connectivity and increased the bandwidth
conservation by limiting the users to traverse multiple inter-switch link into a core
router to make routing decision. Network routing enables a small network to be
linked together with large networks. For example, each classroom has its own
network but whole campus is interconnected and communicating because of network routing. Dynamic routing is best approach to use in large networks because
it automatically detects network changes and updates routing table.
46
REFERENCES
/1/: Hedrick, C. (1988). “Routing Information Protocol”, RFC 1058, Rutgers University. Available in www-form:
<URL: http://tools.ietf.org/rfc/rfc1058.txt>
/2/: Vaasa University of Applied Sciences, General information about VAMK.
Available in www-form:
< URL: http://www.puv.fi/en/prospective_degree_students/why_vamk>
/3/: Malkin, G. (1994). "RIP Version 2 Carrying Additional Information", RFC
1723, Xylogics, Inc. Available in www-form:
<URL: http://www.faqs.org/rfcs/rfc1723.html>
/4/: Ravi, Malhotra. (2002). ”IP Routing”, O'Reilly Media, Inc.
/5/: Puzmanova R.(2002). “Routing and Switching Time of Convergence”, Great
Britain:Addison-Wesley.
/6/: Cisco Exploration: “LAN Switching and Wireless”, 3.1.1 Introducing
VLANs.
/7/: HP Procurve 2910al Manual, Available in www-form:
<URL:
W_14_03.pdf>
http://cdn.procurve.com/training/Manuals/2910-ATG-Feb09-
/8/: Dynamic Host Configuration Protocol, Available in www-form:
<URL: http://technet.microsoft.com/en-us/library/cc781008(WS.10).aspx>
/9/: <URL: http://technet.microsoft.com/en-us/library/cc751462.aspx>
/10/: Layer 3 Switching Network, Available in www-form:
<URL:
http://www.scom.uminho.pt/uploads/Apoio%20-%20Doc%20Tec%20%203switch.pdf>